CN103303205B - Vehicle surroundings monitoring apparatus - Google Patents
Vehicle surroundings monitoring apparatus Download PDFInfo
- Publication number
- CN103303205B CN103303205B CN201310054655.2A CN201310054655A CN103303205B CN 103303205 B CN103303205 B CN 103303205B CN 201310054655 A CN201310054655 A CN 201310054655A CN 103303205 B CN103303205 B CN 103303205B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- moving body
- light
- region
- unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 42
- 230000033001 locomotion Effects 0.000 claims abstract description 43
- 230000005764 inhibitory process Effects 0.000 claims description 22
- 239000013598 vector Substances 0.000 claims description 15
- 238000001514 detection method Methods 0.000 claims description 14
- 238000012360 testing method Methods 0.000 claims description 12
- 230000003287 optical effect Effects 0.000 claims description 11
- 230000033228 biological regulation Effects 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000000034 method Methods 0.000 description 23
- 230000008859 change Effects 0.000 description 17
- 230000008569 process Effects 0.000 description 17
- 238000012545 processing Methods 0.000 description 13
- 238000005286 illumination Methods 0.000 description 8
- 238000012986 modification Methods 0.000 description 6
- 230000004048 modification Effects 0.000 description 6
- 230000003584 silencer Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000003068 static effect Effects 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 230000001678 irradiating effect Effects 0.000 description 3
- 238000012806 monitoring device Methods 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 230000000454 anti-cipatory effect Effects 0.000 description 2
- 210000000746 body region Anatomy 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 235000019504 cigarettes Nutrition 0.000 description 1
- 239000000498 cooling water Substances 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 239000002537 cosmetic Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000002045 lasting effect Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
- G06T2207/30261—Obstacle
Abstract
The invention provides a kind of vehicle surroundings monitoring apparatus, the motion of its outward appearance that can cause with simple structure eliminating aqueous vapor or light source variation, detects the pedestrian that there is the possibility contacted with this car.From the video using in-vehicle camera 111 to take, detect aqueous vapor region based on movable information and monochrome information, make the moving body identification ineffective treatment within a certain period of time in aqueous vapor region.
Description
Technical field
The present invention relates to the vehicle surroundings monitoring apparatus of image detection existence close to the pedestrian of the possibility of this car of the vehicle periphery that basis is taken by in-vehicle camera.
Background technology
In the past, considered the camera on mobile body-mounted 1 more than of vehicle etc., identified the obstacle existed in the surrounding environment of vehicle, as required to chaufeur notice, thus to the system that accident prevents trouble before it happens.Such as, develop according to the multiple image calculation light streams in the shooting of different moment, the motion of each point in synthetic image, the technology of the motion thus in computed image.The known motion by calculating in this image, identifies the image processing apparatus of the moving body such as pedestrian and bicycle.
Such as, in patent documentation 1, describe the light stream in each region of the interior setting of computed image respectively, in the light stream in this each region, based on the light stream more than threshold value preset, identify the technology of the motion of other vehicles.
Patent documentation 1: Japanese Unexamined Patent Publication 6-314340 publication
Summary of the invention
But light stream is subject to the impact of the motion of the outward appearance of the local in image, so brightness value, also at every moment all under the situation of change, may measure the light stream of mistake beyond the moving bodys such as pedestrian.
Such situation, the situation that the aqueous vapor such as contained in emission in the outdoor environment that temperature is lower rises from the exhaust gas silencer of automobile, and may produce when having mirrored shade change that steering indicating light or head lamp cause on road surface etc.That is, even if exist in three dimensional space and there is no obstacle, brightness value along with the time through and under the situation of change, also may measure the problem of the light stream of mistake.
In the vehicle surroundings monitoring apparatus recorded in patent documentation 1, because be identified as other vehicles when light stream is more than defined threshold, so when having measured wrong light stream as mentioned above, even if there are not other vehicles, also vehicle may be mistakenly identified as.
The present invention completes in view of the above problems, and its object is to provides a kind of vehicle surroundings monitoring apparatus, and it can get rid of because the kinetic of the outward appearance in image identifies by mistake, correctly and easily detect the moving body that there is the possibility contacted with this car.
Solve the vehicle surroundings monitoring apparatus of the present invention of above-mentioned problem, based on the multiple image detection moving bodys separating predetermined time interval shooting with in-vehicle camera, the monochrome information based on the movable information of this moving body and image judges the motion of moving body whether outward appearance.Then, there is the region being judged to be the moving body of the motion of outward appearance in shielding, carries out the alarm control corresponding to the testing result of moving body.
According to the present invention, when moving body is the motion of outward appearance, alarm control is carried out in the region that shielding exists this moving body, so when detecting the moving body such as pedestrian's grade that there is the possibility contacted with this car, the error detection that the motion of the outward appearance caused because of aqueous vapor or lights class etc. causes can be prevented.In addition, the explanation of following embodiment of problem other than the above, structure and effect is described.
Accompanying drawing explanation
Fig. 1 is the functional block diagram of the vehicle surroundings monitoring apparatus of present embodiment.
Fig. 2 is the figure of the structure example that vehicle surroundings monitoring apparatus is described.
Fig. 3 is the diagram of circuit of the contents processing that vehicle surroundings monitoring apparatus is described.
Fig. 4 is the figure of the situation represented with rear camera observation water steam.
Fig. 5 represents the figure with the video obtained when the camera observation water steam of rear.
Fig. 6 is the figure representing the situation of irradiating with rear camera observation steering indicating light.
Fig. 7 is the figure representing the video obtained by the situation that rear camera observation steering indicating light irradiates downwards.
Fig. 8 is the figure representing the situation of irradiating with front camera observation head lamp.
The figure of the video that Fig. 9 obtains when being and representing and irradiate with front camera observation head lamp.
Figure 10 is the figure of other structure example that vehicle surroundings monitoring apparatus is described.
Figure 11 is the figure of other structure example that vehicle surroundings monitoring apparatus is described.
Figure 12 is the figure of other structure example that vehicle surroundings monitoring apparatus is described.
Figure 13 is the functional block diagram of other structure example that vehicle surroundings monitoring apparatus is described.
Figure 14 is the functional block diagram that appearance determination unit is described.
Nomenclature
100 vehicle surroundings monitoring apparatus
101 photographic images acquisition unit
102 optical flow computation unit
103 light and shade identifying units
104 moving body detecting units
105 appearance determination unit
106 alarm inhibition zone setup units
107 Alarm control unit
110ECU
111 in-vehicle cameras
Detailed description of the invention
[the first embodiment]
Then, at following explanation first embodiment.
First object of the vehicle surroundings monitoring apparatus of present embodiment is to provide a kind of vehicle surroundings monitoring apparatus, the moving region of its outward appearance caused by detection aqueous vapor or light source variation, make the moving body testing result ineffective treatment in this region, suppress the kinetic alarm by mistake of outward appearance, trouble is not caused to user.
Vehicle surroundings monitoring apparatus is in order to detect the moving body around this car, calculating in-vehicle camera separates the light stream between the photographic images captured by time gap that presets, if have more than the set of pixels unification determined number of unidirectional stream, then export as moving body.
As the aqueous vapor of the major cause of alarm by mistake, there is the feature of following (1) ~ (5).(1) daytime, contrast ratio was lower, (2) when night is irradiated by the head lamp of Ben Che and other cars and steering indicating light etc., brightness uprises and becomes the brightness value of the Near The Upper Limit of imaging apparatus (image pickup device: imageing sensor) output area, (3) rise, spread, be subject to the impact of wind from exhaust gas silencer and flow, (4) motion self of the light stream of aqueous vapor inside is unstable, (5) temperature and damp condition suitable time continue within the specific limits to produce.Wherein, so-called contrast ratio refers in the brightness value of the pixel comprised in certain image-region herein, the difference of the value of its minimax.
In vehicle surroundings monitoring apparatus, the condition such as outside air temperature and engine temperature is judged with onboard sensor, judging whether to be easy is the situation of moving body by aqueous vapor error detection, when there is the possibility of error detection, moving region is extracted by light stream, easily observe the brightness value threshold value of aqueous vapor according to slot setup, judge by the contrast ratio that the threshold value of setting is carried out in the moving region of above-mentioned extraction.When being judged to be the contrast ratio of similar water steam, carry out shielding the moving body making not report in certain hour in this region and start to detect.
Such as, the aqueous vapor on daytime is sighted the white cigarette of light dim low contrast mostly.Therefore, when being judged to be daytime in judging round the clock, when the brightness value being determined with the region of the motion of outward appearance by light stream continuous erratic variation and contrast ratio in regional area more than certain hour are judged to be lower than certain value, judge that this region is the region having photographed aqueous vapor, even if carry out occurring that light stream also suppresses the process reported within a certain period of time in this region.
The aqueous vapor at night, when not being subject to throwing light on, can not detect that the degree of light stream is not easily observed, but when being subject to the illumination of the head lamp of Ben Che and other cars or steering indicating light, street lamp etc., mostly becomes the high brightness of brightness value Near The Upper Limit.So, when being judged to be night in judging round the clock, the brightness value in the region of the motion of outward appearance is judged to be by light stream, when to enter from the upper limit side of the range of luminance values that can obtain the pixel count within 10% and reach more than the area ratio of regulation, judge that this region is aqueous vapor region, even if occur that light stream also suppresses notification within a certain period of time in this region.
In addition, when head lamp and steering indicating light flicker, along with light is irradiated on road surface and wall, there is high-brightness region in image.The Strength Changes of light source slowly, so the border of high-brightness region is moved in appearance, observes light stream compared with the shooting cycle of camera.
Thus, by representing head lamp, steering indicating light, fog lamp, backing-up lamp, the illumination sensor of bright dark state of looking the lights classes such as wide lamp, taillight, registration mark light or the illumination controller of lights class, the lights class learning this car is illuminating state, when light stream areas adjacent exists the high-brightness region reached from the upper limit side of the range of luminance values that can obtain within 10%, suppress the notice that the light stream observed around high-brightness region causes.
And then, for the lamp that steering indicating light so periodically glimmers, by observing its cycle, detect the region of the impact being the steering indicating light flicker being subject to Ben Che or other cars, make the moving body testing result ineffective treatment in this region, suppress the kinetic alarm by mistake of outward appearance.
Wherein, be that example describes by the pixel of the brightness value had from the upper limit side of the range of luminance values that can obtain within 10% as the situation of high-brightness region above, but have according to the brightness regulation aperture of surrounding and the mechanism of shutter speed and automatic gain function, the function of carrying out carrying out when A/D conversion imports video memory the correction of color and brightness to camera video in common camera, so also can be the structure adjusting the range of luminance values being judged to be high-brightness region with these setting values in linkage.
In vehicle surroundings monitoring apparatus, also can use the distribution in the direction of the flow vector of each pixel measured by light stream, carry out mirroring or the judgement of aqueous vapor of lights class.Then, also can consider swept volume and the exhaust gas silencer position of equipped vehicle, the position of the aqueous vapor occurred on picture when supposing windless condition and size, calculate the average of the brightness value in corresponding area size and variance, judge low contrast state.In addition, be judged to be lights class mirror region or aqueous vapor region when, within a certain period of time using the region of this region as non-notify object, again be judged to be within the time of non-notify object lights class mirror region or aqueous vapor region, also can using again within a certain period of time as the mode update time in the region of non-notify object from the moment be determined.
[embodiment 1]
Then, the embodiment of vehicle surroundings monitoring apparatus 100 is described with accompanying drawing.
The present embodiment relates to takes vehicle periphery with in-vehicle camera 111, exports the vehicle surroundings monitoring apparatus 100 of alarm when the moving body of the possibility that there is contact vehicle 120 being detected.
First carry out for illustration of the definition of term.Light stream is called by according to amount of movement association results, the i.e. mobile vector of each image coordinate in 2 image calculation of not taking in the same time with shooting unit.
Due to as aqueous vapor shape itself at every moment all in change, even if so in fact object do not move, also there is the situation of light stream of calculating.In addition, when light illumination environmental turbulence beyond this vehicle or other vehicles, vehicle, gradual change level (gradation) on image and illuminated portion and non-irradiated portion boundary region etc. change, even if so in fact object does not move, also there is the situation calculating light stream.The motion that cosmetic variation in this sampled images causes, is called the motion of outward appearance afterwards.Further, by relative to the shooting viewpoint in actual environment and world coordinate system, the amount of movement of the relative amount of movement of personage or these moving bodys of vehicle hereinafter referred to as object.
The figure of Fig. 1 to be the figure of the structure of the vehicle surroundings monitoring apparatus that present embodiment is described, Fig. 2 be structure example that vehicle surroundings monitoring apparatus is described.
Vehicle surroundings monitoring apparatus 100, for monitoring that pedestrian's this car whether relative moves up in relatively close side, as shown in Figure 2, such as, is formed in the ECU110 of image procossing.Be connected with at the input side of ECU110: for observing the in-vehicle camera 111 of this vehicle-surroundings such as vehicle front or rear; Obtain the vehicle-wheel speed sensor 121 of the rotating speed of each wheel of vehicle; Obtain the rotation angle sensor 122 of the anglec of rotation of bearing circle; For the illumination sensor 123 of the illuminating state of the head lamp and these lights classes of steering indicating light that obtain this vehicle, be connected with the loud speaker 112 for exporting alarm tone and the telltale 113 for the object of display translation alarm at outgoing side.
In-vehicle camera 111 is so-called single-lens cameras (slr camera), is installed on vehicle 120 to take the surrounding of this car.In-vehicle camera 111 is not limited to the rear camera of shooting rear view of vehicle as shown in Figure 4, also can be the front camera of shooting vehicle front and the side camera of shooting vehicle side, also can possess all these cameras.
In addition, vehicle surroundings monitoring apparatus 100 is not limited to and forms in image procossing ECU110, also can form in other vehicle-mounted ECU such as the ECU of special ECU or in-vehicle camera 111, also can consist of the combination of multiple ECU.
Vehicle surroundings monitoring apparatus 100 as shown in Figure 1, has photographic images acquisition unit 101, optical flow computation unit 102, light and shade identifying unit 103, moving body detecting unit 104, appearance determination unit 105, alarm inhibition zone setup unit 106, Alarm control unit 107, information of vehicles acquisition unit 108.
Photographic images acquisition unit 101 obtains the multiple images 1,2 being separated the time gap shooting preset by in-vehicle camera 111.The optical flow computation unit 102 multiple image calculation light streams obtained by photographic images acquisition unit 101.Light and shade identifying unit 103 judges the light and shade of the environment of vehicle periphery based on information of vehicles.What illumination sensor 123 obtained the vehicle lights class of head lamp and lampet etc. lights information.
Moving body detecting unit 104 detects moving body based on light stream.Based on the brightness of light stream and image, appearance determination unit 105 judges that whether moving body is the motion of outward appearance.Setup unit 106 shielding of alarm inhibition zone is judged to be the region that the moving body of the motion of outward appearance exists.Alarm control unit 107 carries out alarm control based on the testing result of the moving body existed in the moving body surveyed area beyond alarm inhibition zone.Information of vehicles acquisition unit 108 obtains information from vehicle-wheel speed sensor 121, rotation angle sensor 122, illumination sensor 123 as information of vehicles.
Then, each structure of vehicle surroundings monitoring apparatus 100 is described in detail.
Vehicle camera 111 is such as the device intensity of the visible ray irradiated the photo-sensitive cell of CCD camera or CMOS camera etc. or the light such as near infrared light or far red light exported as electric charges amplify.
In recent years, also there are built-in memory element and treater in camera, in camera, revise the camera that lens distortion is carried out exporting after photographing.Export and mostly use analog or digital signal, the situation for analog signal in the present embodiment is described.The vision signal exported carries out A/D conversion by photographic images acquisition unit, and the voltage that now also can be converted by change A/D and the mapping parameters of brightness value, obtained the video of same signal with brighter or darker brightness value and preserve in memory.
Preserving like this in image in memory, the 2 width images different with the moment and image 1 and image 2, as input, calculate light stream.
Light stream with being called that the different 2 width images of reference picture and searching image, shooting time are as input, search with reference to certain image block of image and which region similarity of searching image high, be considered as original image block and move to the high region of similarity.This mobile vector is called flow vector, by calculating flow vector respectively to multiple image block, extracts between 2 width images whether domain of the existence moves.
Light and shade identifying unit 103 uses the information of vehicles such as lamp signal, illuminance transducer, clock information of head lamp and fog lamp etc., and whether whether the environment judging vehicle periphery as compared with dark situation or as just in dimmed situation.Then, this result of determination is sent to appearance determination unit 105.
As shown in figure 14, appearance determination unit 105 determines whether to detected the motion of the outward appearance that aqueous vapor or light source variation cause.Appearance determination unit 105 has low contrast regions identifying unit 311, aqueous vapor identifying unit 312, light source variation identifying unit 313.
Low contrast regions identifying unit 311 is for the image 1,2 obtained by photographic images acquisition unit 101, calculate the aviation value of brightness value of pixel or the variation of at least one party of variance yields that comprise in regional area, carry out low contrast judgement based on light and shade determination information.
Iamge Segmentation is the image block of given size by low contrast regions identifying unit 311, calculates the average of the brightness value of the pixel in each piece and variance, according to the result of determination of light and shade identifying unit 103, is switched and determined threshold value.Such as, when result of determination is daytime, in order to the aqueous vapor detecting light aqueous vapor, namely can not clearly see, be low contrast regions by regional determination low for variance yields.Further, when result of determination is night, in order to detect the brighter aqueous vapor thrown light on, be low contrast regions by regional determination high for aviation value.
In aqueous vapor identifying unit 312, based on the result of low contrast identifying unit and optical flow computation unit 102, judge that whether this region is as producing the region having aqueous vapor.Such as, be judged to be night by light and shade identifying unit 103 and only export when having a light stream around the region that brightness value is high, carrying out being defined as the judgement producing and have the region of aqueous vapor.
Whether the brightness value of each regional area in light source variation identifying unit 313 process decision chart picture defers to the pattern (pattern) of regulation along with time variations, determine whether the light source variation that the flicker producing head lamp or steering indicating light causes thus.
Such as, when the head lamp of this car is lighted, trigger by CAN signal or from the lamp lighting start instruction conduct that the hardwire lighting switch generates, the lamp of this car is brightened gradually until the brightness value increase pattern of the timing variations lighted completely stores in the memory unit as anticipatory knowledge.
Then, for the change of the brightness value of each regional area in image, the increment rate of the brightness value that the brightness value before relatively indicating with above-mentioned lamp lighting start is benchmark and brightness value increase the increment rate of pattern, if consistent with each other within the error of defined threshold, are then judged to be that head lamp is lighted.
In addition, about the flicker of the steering indicating light of this car, in the same manner as the lighting of above-mentioned head lamp, using lighting instruction and extinguishing instruction and increase pattern and brightness value as the brightness value of the brightness value triggered and reduce pattern as anticipatory knowledge storage in the memory unit, by comparing the flicker that can judge steering indicating light with the similarity of these patterns with lamp.
In addition, about head lamp or the steering indicating light of other cars, the moment of lamp flicker can not be obtained as information of vehicles.Therefore, such as observation timing variations as described below, infers the cycle of periodicity pattern and compares the similarity with its pattern.
Such as, when steering indicating light, its cycle is defined as flicker in 1 minute 60 times or 120 times, and brightness changing pattern when lighting and when extinguishing is fixed respectively by each vehicle.Although different from vehicle number, but commercially available steering indicating light such as sends lamp and lights sign on, brighten gradually in 200 milliseconds begin to flow through electric current in bulb after, the state continuance of high-high brightness 200 milliseconds, lamp is sent and extinguishes start instruction, self-stopping technology is to completely dimmed in after the electric current supply of bulb 160 milliseconds, and OFF state continues 240 milliseconds.
According to brightness changing pattern when lighting and when extinguishing, the brightness variation model that imagination is parameter with brightness build up time t1, high-high brightness time length t2, brightness t3 fall time, minimum brightness time length t4, according to acquired sequential chart as calculating parameter and the moment of lighting instruction and extinguishing instruction.This moment is by using the existing methods such as method of least square to solve to the brightness change of each regional area in image and brightness variation model.
If brightness changing pattern can be calculated and light instruction and extinguish the moment of instruction, then same with above-mentioned head lamp, by judging whether consistent within certain error luminance transformation pattern changes with the brightness obtained from image, can judge whether to exist the light source variation that steering indicating light causes.Then, when being judged to be the light source variation that steering indicating light causes, be judged to be that moving body is the motion of the outward appearance that light source variation causes.
The result of determination of at least one party in aqueous vapor identifying unit 312 and light source variation identifying unit 313 and the region judged export alarm inhibition zone setup unit by result of determination output unit 315.
The region of the aqueous vapor that aqueous vapor identifying unit exports by alarm inhibition zone setup unit 106 within a certain period of time keeps as alarm inhibition zone, for the moving body newly detected in this region in this time, additional representation newly detects the information of moving body in alarm inhibition zone, suppresses unit 107 notice to be invalid moving body to alarm.In addition, when being judged to be to there is light source variation, using this region as alarm inhibition zone, for the moving body newly detected in this region, additional representation newly detected the information of moving body in alarm inhibition zone, is invalid moving body to Alarm control unit 107 notice.
On the other hand, the moving body detected beyond the region of aqueous vapor or beyond light source variation region moves into these regions, alert objects actv. moving body is existed to Alarm control unit 107 notice.In addition, for the moving body newly detected beyond the region of aqueous vapor or beyond light source variation region, there is new alert objects actv. moving body to Alarm control unit 107 notice.
Alarm control unit 107, based on the result of optical flow computation unit 102 and alarm inhibition zone setup unit 106, only notifies the process of actv. moving body.Alarm control unit 107 controls monitoring device 113 and the loud speaker 112 of Vehicular navigation system and the sound equipment display system arranged in car.Alarm control unit 107, such as carries out the control of Overlapping display alarm display or the control of the sound that gives the alarm from loud speaker in camera video for the output of navigation screen (monitoring device) to user.Further, at least carry out as alarm inhibitory control the control suppressing alarm tone.
The vehicle surroundings monitoring apparatus 100 of the present embodiment, at least by above structure from the shooting moment obtained by vehicle-mounted camera 111 different multiple image contract light streams, determine whether and the area coincidence being judged to be aqueous vapor or light source variation, switch thus and export with or without alarm.
Then, illustrate with Fig. 2 and implement the ECU(Electric Control unit of present treatment: ECU (Electrical Control Unit)) with the interlock of its peripheral device.
ECU110 in order to calculate this garage now move from camera 111 input video, and receives the sensor information from vehicle-wheel speed sensor 121 and rotation angle sensor 122.As long as this sensor moves and the sensor used to calculate this garage, car speed sensor, wheel speed pulses sensor, steering wheel angle sensor, rudder angle power assist apparatus, vehicle height sensor, yaw rate sensor, GPS sensor, acceleration pick-up etc. can be used.In addition, illumination sensor 123 is sensors of the state of the lights class representing this car, and such as, can judge around under the situation of pre-irradiation illuminator is darker environment.In addition the illuminance transducer etc. used in automatic headlamp lamp device can also be used.
ECU110 shows the result performing vehicle periphery and monitor at telltale 113, to give the alarm sound as required in order to arouse the attention of chaufeur from loud speaker 112.
Diagram of circuit based on Fig. 3 illustrates the contents processing with the vehicle surroundings monitoring apparatus of said structure.First, in step slo, with in-vehicle camera 111 to comprise aqueous vapor, road surface vehicle periphery with predetermined time interval shooting at least 2 times, obtain 2 width images 1,2, proceed to step S20.
In step S20, calculate light stream by optical flow computation unit 102 according to 2 width images 1,2, proceed to step S30.In step s 30, detected the moving body be present on road surface by moving body detecting unit 104, proceed to step S40.
In step s 40, the motion of the outward appearance caused by aqueous vapor or light source variation is judged by appearance determination unit 105.In step s 50, alarm inhibition zone is set by alarm inhibition zone setup unit 106.In step S60, judge newly to detect whether the region of moving body repeats with alarm inhibition zone by moving body detecting unit 104.Then, when being judged as repetition, judging that the moving body detected is the kinetic error detection of outward appearance and does not export alarm, proceed to the no path in step S80(figure).On the other hand, under being judged as unduplicated situation, being judged as that the moving body detected is not the kinetic error detection of outward appearance, proceeding to the path being in step S70(figure).
In step S70, export alarm from monitoring device 113 and loud speaker 112, the chaufeur of prompting vehicle is noted.Proceed to step S80 afterwards.In step S80, detect when making the operating switch of monitor unit work around (not shown) cut-off or when the ignition lock of vehicle is cut off, judge that process terminates.At the end of not being judged as, return step S10, repeatedly identical process.
Then, the particular content of the optical flow computation process carried out in step S20 is described.
If the image taken at moment t by camera 111 is It(x, y), the image taken at moment t+ Δ t is It+ Δ t(x, y).First, from image It(x, y) in the larger point of sensed luminance gradient as unique point.Specifically, for image It(x, y), near the pixel paid close attention to, set zonule, make the edge strength obtaining set inside, zonule as the operator effect of the amount of expression brightness step, the edge strength obtained is greater than the pixel of specified value as unique point.Now also can calculate the edge direction in same pixel in the lump.
Then, from image It+ Δ t(x, y) in search with from image It(x, y) in the unique point that detects there is the pixel (corresponding point) of same brightness gradient.This process is by image It+ Δ t(x, y) search coverage of setting prescribed level in, in set search coverage, search many with from image It(x, y) in the unique point that detects there is the pixel of same brightness gradient (edge strength and edge direction) and carry out.
Then, respectively threshold value is arranged for the degree of approximation of edge strength and the degree of approximation of edge direction, when the difference of edge strength and the difference of edge direction are all within the threshold value set respectively, be judged to have found corresponding point.The detection of other unique points is carried out when not searching corresponding point.
Then, determine with from image It(x, y) in the unique point that detects for starting point, with from image It+ Δ t(x, y) in the corresponding point that find be the light stream of terminal.Start position coordinate and the final position coordinate of the light stream detected like this are stored in optical flow computation unit.
For image It(x, y) all pixels carry out the check processing of above-mentioned unique point.Wherein, the method for calculating of light stream is not limited to the above.That is, the method for inspection of light stream has proposed multiple, and any one known method can be used to perform.
The content of the moving body check processing carried out in step S30 is described.
In moving body check processing, read the starting point coordinate of the stream obtained from the result of calculation of light stream, middle point coordinate, stream length, carry out the grouping of flow vector.This process is the process light stream gone out in close position probing being carried out integrating.Specifically, the light stream in the region of the size preset is compared each other, length be more than specified value and the difference in direction below specified value time, light stream is divided into one group.Then, above-mentioned grouping is carried out for all light stream vectors on image.Afterwards, the light stream after being grouped has when prescribed level and is judged to be object on picture.
By the content of the motion determination processing of outward appearance of carrying out in Fig. 4 and Fig. 5 description of step S40.
First, the example of the aqueous vapor detection of one of motion representing outward appearance.
Fig. 4 represents the example detecting aqueous vapor 130 with in-vehicle camera 111.Since it is known the aqueous vapor of this car produces near silencer 131, so can predict from the position the picture corresponding with silencer position as the general knowledge of every chassis and produce aqueous vapor.
Aqueous vapor 130 discharges diffusion from the silencer 131 of vehicle, as shown in Figure 5, mirrors in the image of in-vehicle camera 111 shooting.In the check processing of aqueous vapor, roughly the same position will be positioned in the world coordinate system of observation but the direction of the flow vector of the light stream flow vector of getting the same coordinate on various direction, image is recorded as the flow vector of aqueous vapor as the point of sequential change.This process is carried out for all light stream vectors on image.
Afterwards, divided into groups by the flow vector of the aqueous vapor to record, calculate aqueous vapor region.This integration specifically, be the starting point coordinate of the stream of the flow vector record compared as aqueous vapor, terminal point coordinate whether in the region of the size preset, divide into groups when being in neighbouring distance.Flow vector for all aqueous vapors on image carries out this packet transaction.The output of the moving region of outward appearance, such as by the moving region of each outward appearance obtained by packet transaction, the kind of the motion of the area of the moving region of the outward appearance in the upper left point coordinate of record boundary rectangle and lower-right most point coordinate, boundary rectangle or area ratio, outward appearance.
The content of the setting process of the alarm inhibition zone of carrying out in step S50 is described.
In the setting process of alarm inhibition zone, the motion identifying unit record by outward appearance, the boundary rectangle of the moving region of outward appearance on picture continues to deposit in case in identical place at specified time, is alarm inhibition zone by this regional record.
Whether exist in identical place, specifically, when moment t-1 and moment t obtains upper left point coordinate and the lower-right most point coordinate of multiple boundary rectangle, the coincidence factor in its region is more than the threshold value specified, be judged to be present in identical place.When obtaining boundary rectangle from moment t-N in the N+1 of moment t not image in the same time, if it is determined that be present in identical place for more than regulation number, be then judged to be present in identical place at N+1 time remaining.
Then, the light source variation of one of its motion as his outward appearance is described by Fig. 6 and Fig. 7.
Fig. 6 represents the figure making the situation on light source irradiation road surface because of the flicker of steering indicating light, and Fig. 7 is the figure representing the situation that irradiated road surface is observed as image.
Stationary vehicle and road surface evenness when, position and the projecting pattern of the light source 140 of phase road pavement are fixed, thus with camera 111 observe road surface on brightness change position 141 also fix.In addition, when the light source scintillation that this static car or other static cars carry on this road surface, the fixed area on image is subject to the impact of light source variation and produces brightness change.And then, not all regions on road surface, but only have the fixed area corresponding to the shape of light source to be subject to the impact of light source variation.Therefore, be regional area by Iamge Segmentation, observe the series brightness in divided region change.
The observation of the brightness change of sequential is as described below to be carried out.First, according to when lighting and extinguish time brightness changing pattern imagination using brightness build up time t1, high-high brightness time length t2, brightness t3 fall time, minimum brightness time length t4 as the brightness variation model of parameter, calculate this parameter.
Brightness value aviation value is calculated respectively for each regional area in the image of acquired sequential, in the Circular buffer with the spread length being equivalent at least longer than 1 cycle time (such as 1.5 cycles), preserves savings.Preserve the average brightness longer than 1.5 cycles in Circular buffer after, arrange the part with same waveform shape with the brightness value in 0.5 cycle of up-to-date side, computation period thus from Circular buffer search.
Now, imagination sampling frequency changes and insufficient short situation relative to brightness value, also can generate curve of approximation according to the brightness value sequence in 0.5 cycle, search for this curve of approximation similar portion each other.
Curve of approximation is herein 3 curves, 4 curves etc., when obtaining multiple average brightness and observation moment thereof, easily with the curve of the approximate generations such as method of least square by brightness value m-time this.
By searching for the phase place of the curve of approximation of this generation, the cycle can not be calculated by the impact in sampling interval.Calculate all after dates, calculate maximum brightness value time length t2 and minimum luminance value time length t4.This is because brightness value is roughly fixing during t2, t4, so for the brightness value sequence comprised in Circular buffer, calculate the average and brightness variance of the brightness of specified time length, be judged to be that when variance yields is less than threshold value t2, t4 are lasting, calculate thus.T1, t3 can calculate with removing the time of t2, t4 from cycle entirety.
About above Fig. 6 and Fig. 7 described, the situation of light source scintillation contemplating the road surface that there is not stereoscopic article and on this road surface, this static car or other static cars carry.Fig. 8 and Fig. 9 represents to there is wall near this car, the figure of front and then example close to wall while irradiating with this car head lamp.
In Fig. 8, this car 150 is to the light of wall 153 pre-irradiation illuminator, and the range of exposures 152 of this light is observed as the region that brightness value is very high in the picture.
This range of exposures 152 is with proportional to the distance of wall 153 from this car 150, and by this car 150, close to wall 153, range of exposures narrows.Therefore, in the captured image of Fig. 9, produce the motion of outward appearance on the border of irradiation area 152, observe light stream.So, only when existing to the amount of movement of the camera boresight direction of this car 150, the region with the edge of regulation range of luminance values to be extracted as the moving region of outward appearance when projection light in this car 150 pairs of camera fields of view.
In addition, the amount of movement existed to camera boresight direction refers to, the speed change gear such as processing shifter bar when the image of front camera is D(Drive) or L(Low) and detect the state of the speed of a motor vehicle of regulation, or the state of the information that forward direction rotates is obtained from wheel rotation sensor, in addition, the speed change gear such as processing shifter bar when the image of rear camera is R(Reverse) and the state of pulse is obtained from wheel speed pulses sensor, can be judged by the combination of the camera selection mode of system and vehicle sensors.
When there is the amount of movement to camera boresight direction, because the intensity of the light source of lift-launch can be learnt from camera selection mode, so change the intensity at the edge that will delete of setting in advance.For input picture, only extract the edge of this edge strength, calculate the moving region of outward appearance.
As described above, according to the vehicle surroundings monitoring apparatus 100 of the present embodiment, because based on the image that camera 111 is taken, detect the moving region of the outward appearance that aqueous vapor or light source variation cause and make the moving body testing result ineffective treatment in this region, so can provide a kind of and suppress the kinetic alarm by mistake of outward appearance, do not cause troublesome vehicle periphery detecting device to user.
Then, the structure of the hardware carrying this algorithm is described with Figure 10.
Vision signal from multiple stage cameras 161 such as front camera, rear camera, side camera, indoor cameras is kept in memory device 162.CPU160, based on the vision signal of preserving in memory device 162 and sensor information from onboard sensor 169, detects moving body and detects the moving region of outward appearance.Then, in order to notify this result to user, suppressing the kinetic by mistake alarm of outward appearance, selecting suitable camera video, telltale 163 shows, and to give the alarm sound from loud speaker 164.
In addition, from the view point of being carried out final confirmation by user, also can regardless of the testing result of the moving region of outward appearance, moving body testing result is all directly shown to telltale 163, only the alarm tone from loud speaker 164 is carried out to the suppression of alarm tone according to the testing result of the moving region of outward appearance.
Then, from the view point of the flowing of information, be described with Figure 13.
From illuminance transducer, external temperature sensor, exhaust gas temperature sensor/cooling-water temperature sensor (all not shown) etc., appearance determination unit 211 is to represent that the information of the sensor class of engine condition is as input, parameter or the switching of changing aqueous vapor check processing implement/are not implemented.In addition, using the signal lighting instruction of illuminance transducer, lights class as input, the enforcement/do not implement of the motion determination processing of the outward appearance with light source variation is switched.In addition, utilize vehicular movement information, the background calculated with the movement of this car moves.
Camera video as input, is calculated moving region by optical flow computation unit 212 from picture, and the background that removing is calculated by appearance determination unit 211 moves.Afterwards, object detection unit 213 extracts moving body region from inhibit the light stream of background movement.
Alarm inhibition zone setup unit 214 preserves the such as aqueous vapor region obtained by appearance determination unit 211.In alarm identifying unit 215, when detecting moving body region in this aqueous vapor region, suppress notice.
[the second embodiment]
Then, with Figure 11, the second embodiment is described.
The roughly flow process of the process in present embodiment is substantially identical with the first embodiment, so only describe different parts.Figure 11 is the figure of the hardware configuration representing the second embodiment.
In this second embodiment, video modification unit 165 has the structure of the video input receiving camera 161.Video modification unit 165 is such as integrated by overlooking conversion all around respectively to the video of the camera carried from vehicle, generates thus and overlooks monitoring picture.Video modification unit 165 will comprise the generated transmission of video overlooking monitoring picture to CPU160.CPU160 carries out motion detection and the moving body detection of outward appearance.
Motion for the outward appearance of overlooking monitoring picture detects, and for common camera video, namely directly the motion of the outward appearance of video (untreated video) detects substantially identical.The image recognition result performed by CPU160 is transferred to video modification unit 165.
Afterwards, in video modification unit 165, comprising overlapping drawing processing result on the video overlooking monitoring picture, telltale 163 is exported, and the process of the sound that carries out giving the alarm from loud speaker 164.
In addition, as described in the first embodiment, from the view point of being carried out final confirmation by user, also can regardless of the testing result of the moving region of outward appearance, moving body testing result is all directly shown for telltale 163, only suppresses the alarm tone from loud speaker 164 according to the testing result of the moving region of outward appearance.
In addition, such as, by revising lens distortion, more suitable optical flow computation can be carried out, particularly in the region that the distortion such as picture end easily become large, the correctness that the motion expecting to increase zonule detects.
[the 3rd embodiment]
Then, with Figure 12, the 3rd embodiment is described.
The roughly flow process of the process in present embodiment is substantially identical with the first embodiment, so only describe different parts.Figure 12 is the figure of the hardware configuration representing the 3rd embodiment.
In present embodiment, different in the input this point directly receiving camera 161 at CPU160.Thereby, it is possible to reduction bus load, and device can be obtained can the advantage of miniaturization.
More than described embodiments of the present invention in detail, but the present invention is not limited to above-mentioned embodiment, in the scope not departing from the spirit of the present invention recorded in claim, can various design modification be carried out.Such as, above-mentioned embodiment is described in detail for ease of the present invention being described with understanding, and is not limited to possess illustrated all structures.In addition, a part for the structure of certain embodiment can be replaced into the structure of other embodiments, or in the structure of certain embodiment, add the structure of other embodiments.And, for a part for the structure of each embodiment, can add, delete, replace other structures.
Claims (5)
1. a vehicle surroundings monitoring apparatus, it carries out the supervision of vehicle periphery based on the multiple images being separated predetermined time interval shooting by in-vehicle camera, and the feature of this vehicle surroundings monitoring apparatus is, comprising:
Obtain the image acquisition unit of described multiple image;
Based on the moving body detecting unit of the plurality of image detection moving body;
Described moving body is judged whether as the appearance determination unit of the motion of outward appearance based on the movable information of this moving body and monochrome information;
The alarm inhibition zone setup unit of setting alarm inhibition zone, this alarm inhibition zone is shielded the region existing for the moving body being judged to be described apparent motion;
Testing result based on described alarm inhibition zone and described moving body carries out the Alarm control unit of alarm control;
Based on the optical flow computation unit of described multiple image calculation light stream; With
Judge the light and shade identifying unit of the light and shade of vehicle-periphery,
Wherein, described appearance determination unit,
Use the variance yields of the vector components in regulation region of described light stream as described movable information,
Use the aviation value of the brightness value in regulation region of described multiple image and at least one party of variance yields as described monochrome information,
The aviation value of described brightness value or the threshold value of variance yields is switched according to the result of determination of described light and shade.
2. vehicle surroundings monitoring apparatus as claimed in claim 1, is characterized in that:
Described appearance determination unit has:
Low contrast regions identifying unit, region low for the variance yields of described brightness value exports as low contrast regions when judging that vehicle-periphery becomes clear by described light and shade identifying unit by it, on the other hand, when judging that vehicle-periphery is dark, region high for the aviation value of described brightness value is exported as low contrast regions; With
Aqueous vapor identifying unit, it judges described moving body whether as aqueous vapor based on this low contrast regions and described light stream.
3. vehicle surroundings monitoring apparatus as claimed in claim 2, is characterized in that:
Described aqueous vapor identifying unit, is being judged that by described light and shade identifying unit described vehicle-periphery is dark, only when the surrounding in the high region of the aviation value of brightness value export have light stream, be judged to be that described moving body is aqueous vapor.
4. vehicle surroundings monitoring apparatus as claimed in claim 2, is characterized in that:
By described light and shade identifying unit, described aqueous vapor identifying unit, is judging that described vehicle-periphery becomes clear, only export when having a light stream around the region that the variance yields of brightness value is low, be judged to be that described moving body is aqueous vapor.
5. vehicle surroundings monitoring apparatus as claimed in claim 1, is characterized in that:
Described appearance determination unit has:
Light source variation identifying unit, the brightness value of its multiple regional areas set in the picture defers to the pattern of regulation along with time variations when, is judged to there occurs the light source variation that light source scintillation causes,
Be judged to there occurs light source variation by this light source variation identifying unit, judge the motion of the outward appearance that described moving body causes as light source scintillation thus.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2012050908A JP5615862B2 (en) | 2012-03-07 | 2012-03-07 | Vehicle perimeter monitoring device |
| JP2012-050908 | 2012-03-07 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103303205A CN103303205A (en) | 2013-09-18 |
| CN103303205B true CN103303205B (en) | 2015-09-09 |
Family
ID=47748472
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310054655.2A Expired - Fee Related CN103303205B (en) | 2012-03-07 | 2013-02-20 | Vehicle surroundings monitoring apparatus |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US20130235201A1 (en) |
| EP (1) | EP2637150B1 (en) |
| JP (1) | JP5615862B2 (en) |
| CN (1) | CN103303205B (en) |
Families Citing this family (41)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10638221B2 (en) | 2012-11-13 | 2020-04-28 | Adobe Inc. | Time interval sound alignment |
| US9355649B2 (en) | 2012-11-13 | 2016-05-31 | Adobe Systems Incorporated | Sound alignment using timing information |
| US10249321B2 (en) | 2012-11-20 | 2019-04-02 | Adobe Inc. | Sound rate modification |
| US9025822B2 (en) | 2013-03-11 | 2015-05-05 | Adobe Systems Incorporated | Spatially coherent nearest neighbor fields |
| US9129399B2 (en) | 2013-03-11 | 2015-09-08 | Adobe Systems Incorporated | Optical flow with nearest neighbor field fusion |
| US9031345B2 (en) * | 2013-03-11 | 2015-05-12 | Adobe Systems Incorporated | Optical flow accounting for image haze |
| US9165373B2 (en) | 2013-03-11 | 2015-10-20 | Adobe Systems Incorporated | Statistics of nearest neighbor fields |
| JP5906224B2 (en) * | 2013-09-27 | 2016-04-20 | 富士重工業株式会社 | Outside environment recognition device |
| KR102159359B1 (en) * | 2014-09-05 | 2020-09-23 | 현대모비스 주식회사 | Around view monitoring system and the operating method |
| JP6330383B2 (en) * | 2014-03-12 | 2018-05-30 | 株式会社デンソー | Composite image generation apparatus and composite image generation program |
| CN106415598B (en) | 2014-05-27 | 2020-08-28 | 罗伯特·博世有限公司 | Lens contamination detection, identification and mitigation for vehicle-mounted camera systems |
| US10252663B2 (en) * | 2014-06-02 | 2019-04-09 | Denso Corporation | Headlight control apparatus |
| JP6314666B2 (en) * | 2014-06-02 | 2018-04-25 | 株式会社デンソー | Headlight control device |
| JP6412345B2 (en) * | 2014-06-12 | 2018-10-24 | 株式会社Ｓｕｂａｒｕ | Outside environment recognition device |
| KR101637882B1 (en) * | 2014-08-07 | 2016-07-08 | 엘지전자 주식회사 | Apparatus for driving headlamp of vehicle and vehicle including the same |
| KR102339512B1 (en) | 2014-10-17 | 2021-12-16 | 현대모비스 주식회사 | Apparatus and method for vision recognition |
| DE102014118035B4 (en) * | 2014-12-05 | 2022-11-17 | Valeo Schalter Und Sensoren Gmbh | Method for detecting shadowing of a sensor device of a motor vehicle by an object, computing device, driver assistance system and motor vehicle |
| CN107113375B (en) * | 2015-01-08 | 2020-09-18 | 索尼半导体解决方案公司 | Image processing apparatus, imaging apparatus, and image processing method |
| JP6236039B2 (en) * | 2015-06-26 | 2017-11-22 | 株式会社Ｓｕｂａｒｕ | Outside environment recognition device |
| JP6523196B2 (en) * | 2016-03-17 | 2019-05-29 | 株式会社東芝 | Estimation apparatus, method and program |
| CN106203272B (en) * | 2016-06-29 | 2018-07-20 | 上海小蚁科技有限公司 | The method and apparatus for determining the movement of movable objects |
| WO2018006023A1 (en) * | 2016-06-30 | 2018-01-04 | Referentia Systems, Inc. | Time series data query engine |
| JP6535302B2 (en) | 2016-07-13 | 2019-06-26 | 株式会社Ｓｏｋｅｎ | Object detection device |
| CN107650785A (en) * | 2016-07-26 | 2018-02-02 | 奥迪股份公司 | Driving assistance method and system |
| CN106295583B (en) * | 2016-08-15 | 2020-02-07 | 深圳市华宝电子科技有限公司 | Method and device for reminding user of driving mode |
| US10726278B2 (en) * | 2016-09-30 | 2020-07-28 | Samsung Electronics Co., Ltd. | Method, device and system for providing notification information |
| KR20180069147A (en) * | 2016-12-14 | 2018-06-25 | 만도헬라일렉트로닉스(주) | Apparatus for warning pedestrian in vehicle |
| JP6852534B2 (en) * | 2017-04-12 | 2021-03-31 | アイシン精機株式会社 | Obstacle detection and notification device, method and program |
| US10549853B2 (en) | 2017-05-26 | 2020-02-04 | The Boeing Company | Apparatus, system, and method for determining an object's location in image video data |
| JP6327388B2 (en) * | 2017-06-15 | 2018-05-23 | 株式会社Ｊｖｃケンウッド | Captured image display device, captured image display method, and captured image display program |
| JP6311826B2 (en) * | 2017-06-15 | 2018-04-18 | 株式会社Ｊｖｃケンウッド | Captured image display device, captured image display method, and captured image display program |
| US10789682B2 (en) * | 2017-06-16 | 2020-09-29 | The Boeing Company | Apparatus, system, and method for enhancing an image |
| JP6878221B2 (en) * | 2017-09-12 | 2021-05-26 | 日立建機株式会社 | Obstacle detection system for work machines |
| JP2019204193A (en) * | 2018-05-22 | 2019-11-28 | キヤノン株式会社 | Image processing device, image processing method, and program |
| JP7279721B2 (en) * | 2018-10-30 | 2023-05-23 | 日本電気株式会社 | Object recognition device, object recognition method and object recognition program |
| JP7252755B2 (en) * | 2018-12-27 | 2023-04-05 | 株式会社小糸製作所 | Active sensors, object identification systems, vehicles, vehicle lighting |
| JP7298436B2 (en) | 2019-10-14 | 2023-06-27 | 株式会社デンソー | Obstacle identification device and obstacle identification program |
| CN112235513B (en) * | 2020-09-29 | 2023-04-07 | 三一专用汽车有限责任公司 | Image processing device, method and vehicle |
| CN117121075A (en) * | 2021-03-03 | 2023-11-24 | 日产自动车株式会社 | Object detection method and object detection device |
| JP2023014606A (en) * | 2021-07-19 | 2023-01-31 | 株式会社小糸製作所 | Control device of vehicle |
| CN115690162B (en) * | 2022-12-28 | 2023-03-10 | 武汉凡德智能科技有限公司 | Method and device for detecting moving large target in fixed video |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06290255A (en) * | 1993-03-31 | 1994-10-18 | Ohkura Electric Co Ltd | Moving vector detecting method for image |
| GB2282294A (en) * | 1993-09-25 | 1995-03-29 | Bosch Gmbh Robert | Method of detecting moving objects |
| CN1457023A (en) * | 2002-05-09 | 2003-11-19 | 松下电器产业株式会社 | Monitor, monitoring method and programm |
| EP1962254A2 (en) * | 2007-02-13 | 2008-08-27 | Hitachi, Ltd. | In-vehicle apparatus for recognizing running environment of vehicle |
| JP4333462B2 (en) * | 2004-04-20 | 2009-09-16 | 株式会社デンソー | Moving object detection device |
| WO2011016367A1 (en) * | 2009-08-04 | 2011-02-10 | アイシン精機株式会社 | Vehicle-surroundings awareness support device |
| JP2011123613A (en) * | 2009-12-09 | 2011-06-23 | Fuji Heavy Ind Ltd | Stop line recognition device |
| CN102257533A (en) * | 2008-11-19 | 2011-11-23 | 歌乐牌株式会社 | Approaching object detection system |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2994170B2 (en) | 1993-04-30 | 1999-12-27 | 三菱電機株式会社 | Vehicle periphery monitoring device |
| JPH07262375A (en) * | 1994-03-25 | 1995-10-13 | Toshiba Corp | Mobile object detector |
| WO2001033503A1 (en) * | 1999-11-03 | 2001-05-10 | Cet Technologies Pte Ltd | Image processing techniques for a video based traffic monitoring system and methods therefor |
| CN101023407A (en) * | 2002-08-21 | 2007-08-22 | 金泰克斯公司 | Image acquisition and processing methods for automatic vehicular exterior lighting control |
| JP3987013B2 (en) * | 2003-09-01 | 2007-10-03 | 本田技研工業株式会社 | Vehicle periphery monitoring device |
| US7389171B2 (en) * | 2003-12-22 | 2008-06-17 | Ford Global Technologies Llc | Single vision sensor object detection system |
| US20060111841A1 (en) * | 2004-11-19 | 2006-05-25 | Jiun-Yuan Tseng | Method and apparatus for obstacle avoidance with camera vision |
| JP4956374B2 (en) * | 2007-10-29 | 2012-06-20 | 富士重工業株式会社 | Object detection device and contact avoidance system |
| JP2012243049A (en) * | 2011-05-19 | 2012-12-10 | Fuji Heavy Ind Ltd | Environment recognition device and environment recognition method |
-
2012
- 2012-03-07 JP JP2012050908A patent/JP5615862B2/en active Active
-
2013
- 2013-02-19 US US13/770,159 patent/US20130235201A1/en not_active Abandoned
- 2013-02-19 EP EP13155832.2A patent/EP2637150B1/en not_active Not-in-force
- 2013-02-20 CN CN201310054655.2A patent/CN103303205B/en not_active Expired - Fee Related
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06290255A (en) * | 1993-03-31 | 1994-10-18 | Ohkura Electric Co Ltd | Moving vector detecting method for image |
| GB2282294A (en) * | 1993-09-25 | 1995-03-29 | Bosch Gmbh Robert | Method of detecting moving objects |
| CN1457023A (en) * | 2002-05-09 | 2003-11-19 | 松下电器产业株式会社 | Monitor, monitoring method and programm |
| JP4333462B2 (en) * | 2004-04-20 | 2009-09-16 | 株式会社デンソー | Moving object detection device |
| EP1962254A2 (en) * | 2007-02-13 | 2008-08-27 | Hitachi, Ltd. | In-vehicle apparatus for recognizing running environment of vehicle |
| CN102257533A (en) * | 2008-11-19 | 2011-11-23 | 歌乐牌株式会社 | Approaching object detection system |
| WO2011016367A1 (en) * | 2009-08-04 | 2011-02-10 | アイシン精機株式会社 | Vehicle-surroundings awareness support device |
| JP2011123613A (en) * | 2009-12-09 | 2011-06-23 | Fuji Heavy Ind Ltd | Stop line recognition device |
Also Published As
| Publication number | Publication date |
|---|---|
| JP5615862B2 (en) | 2014-10-29 |
| JP2013186668A (en) | 2013-09-19 |
| US20130235201A1 (en) | 2013-09-12 |
| CN103303205A (en) | 2013-09-18 |
| EP2637150A2 (en) | 2013-09-11 |
| EP2637150B1 (en) | 2015-04-01 |
| EP2637150A3 (en) | 2014-02-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103303205B (en) | Vehicle surroundings monitoring apparatus | |
| KR102135427B1 (en) | Systems and methods for processing streamed video images to correct flicker of amplitude-modulated light | |
| JP6132412B2 (en) | Outside environment recognition device | |
| CN102971776B (en) | Obstacle detection system and method, obstacle detector | |
| CN103249597A (en) | Vehicle light distribution control device and method | |
| JP2011178301A (en) | Obstacle detection device, obstacle detection system including the same, and obstacle detection method | |
| KR101738995B1 (en) | Imaging system and method with ego motion detection | |
| CN113246846B (en) | Vehicle light control method and device and vehicle | |
| CN106915295B (en) | The control method and device of automobile front lamp state | |
| CN106458084A (en) | Vehicular lighting device | |
| JP6236039B2 (en) | Outside environment recognition device | |
| JP2009071789A (en) | Vehicle periphery monitoring system | |
| JP4927647B2 (en) | Vehicle periphery monitoring device | |
| CN105034929A (en) | Irradiation apparatus | |
| JP2016110539A (en) | Vehicle-outside environment recognition device | |
| CN104008518B (en) | Body detection device | |
| JP4731177B2 (en) | Infrared imaging display device and infrared imaging display method for vehicle | |
| JP5712821B2 (en) | Shooting display control system | |
| JP2013235444A (en) | Vehicle view support apparatus | |
| CN110588499A (en) | Machine vision-based headlamp adaptive control system and method | |
| CN110399672B (en) | Street view simulation method and device for unmanned vehicle and electronic equipment | |
| JP2006252363A (en) | Detector for object around vehicle, and method of detecting object around vehicle | |
| JP2000180563A (en) | Rainfall/snowfall condition detection method and device | |
| KR20160091331A (en) | System and method for forming nighttime images for a motor vehicle | |
| CN112565618B (en) | Exposure control device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150909 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |